General-Purpose Image Forensics Using Patch Likelihood under Image Statistical Models
|
|
- Spencer Dixon
- 5 years ago
- Views:
Transcription
1 General-Purpose Image Forensics Using Patch Likelihood under Image Statistical Models Wei Fan, Kai Wang, and François Cayre GIPSA-lab, CNRS UMR5216, Grenoble INP, 11 rue des Mathématiques, F St-Martin d Hères Cedex, France {wei.fan, kai.wang, francois.cayre}@gipsa-lab.grenoble-inp.fr Abstract This paper proposes a new, conceptually simple and effective forensic method to address both the generality and the fine-grained tampering localization problems of image forensics. Corresponding to each kind of image operation, a rich GMM (Gaussian Mixture Model) is learned as the image statistical model for small image patches. Thereafter, the binary classification problem, whether a given image block has been previously processed, can be solved by comparing the average patch log-likelihood values calculated on overlapping image patches under different GMMs of original and processed images. With comparisons to a powerful steganalytic feature, experimental results demonstrate the efficiency of the proposed method, for multiple image operations, on whole images and small blocks. Index Terms General-purpose image forensics, fine-grained tampering localization, natural image statistics, Gaussian mixture model, patch likelihood I. INTRODUCTION During the last a few years, image forensics has become a commonly acknowledged image authentication tool to expose doctored images in a blind and passive way. Since image forgery creation usually involves various image processing techniques, lots of efforts in the image forensics community have been put to tracing image processing history. Consequently, a large number of forensic algorithms were proposed to reveal different image operations, here to mention a few, e.g., JPEG compression [1], median filtering [2], and resampling [3]. Most of these techniques target at identifying a specific image operation, and are indeed powerful at detecting images processed by the image operation under investigation. They in literature are labelled as targeted (ad hoc) schemes, which however are known to be lack of generality. For example, a powerful forensic feature designed for identifying JPEG compression may not be effective enough to build a median filtering detector. In order to overcome the generality limitation of targeted forensics, general-purpose image forensics has recently emerged to cope with multiple image operations. Though general-purpose image forensics appears to be a more practical (also more challenging-to-design) technique than targeted forensics, its development is still at a very early stage. Probably the only existing work addressing the generality issue of image forensics is the paper of Qiu et al. s [4]. In their forensic strategy, steganography is regarded as a kind of image processing. Therefore, powerful steganalytic features, such as SPAM (Subtractive Pixel Adjacency Matrix) [5], SRM (Spatial-domain Rich Model) [6], and LBP (Local Binary Patterns) [7], are used to build forensic detectors exposing different image operations. In fact, adopting steganalytic features for image forensic purposes is not completely new. Kirchner et al. [8] have previously showed the effectiveness of the SPAM feature [5] to serve for median filtering forensics. For JPEG forensic purposes, Li et al. [9] also borrowed a feature [10] from steganalysis. Modern image steganalysis has been developing to successfully detect stego-images with low-rate embedded data by powerful steganography methods. This is realized by the design of complex high-dimensional steganalytic features and the adoption of machine learning methods, such as the ensemble classifier [11] and the SVM (Support Vector Machine) [12]. It is well demonstrated in [4] that these steganalytic features successfully capture the image statistical change caused by different image operations, which actually produce a much higher image modification rate than steganography. The image forensic methods based on steganalytic features [4], [8], [9] as well as a lot of other image forensic algorithms only report, indeed with a very high accuracy, whether a given image of big size has been processed by a certain image operation. However, no results have been provided in [4], [8], [9] concerning the forensic effectiveness of steganalytic features on small image blocks 1, which can be taken as an equivalence to image tampering localization. Though a simple binary answer may be considered sufficient for steganalysis (linked to Simmons prisoner problem [13]), it may not be informative enough for image forensics to help people grasp a deep understanding of the image tampering locations and semantics. In this paper, we are particularly interested in exposing the image operation locally in the image. This is due to the common practice of using cut-and-paste or copy-move to create image forgeries with modified semantics. In this case, it frequently happens that only certain local image areas rather than the whole image have undergone a particular image processing. For revealing the semantics of the tampered image, image forensics should be able to expose image processing on small image blocks. Based on the above discussion, we would like to emphasize 1 In this paper, we use two similar words block and patch with subtle difference. Unless otherwise addressed, block is used to referred to a relatively big image area, rectangle but not necessarily square, and blocks do not overlap. However, patch is usually a small square image area, and patches can overlap.
2 the importance of both the generality and the ability to perform fine-grained tampering localization for image forensics. With these two goals in mind and following a very different strategy than that of the general-purpose image forensic method [4] inherited from steganalysis, in this paper, we propose a new, conceptually simple and effective way to conduct image forensics based on likelihood comparison under parametric image statistical models. The underlying intuition of the proposed method is reflected by answering the following question: given an image block, is it more like a natural, original block or a processed one? In order to answer this question, for each image operation in consideration (or no processing at all), we learn an image statistical model using the GMM (Gaussian Mixture Model) on small image patches. Thereafter, we are able to measure how likely a new given image block has been previously processed by a certain image operation, by calculating its average patch log-likelihood value under the corresponding GMM. Meanwhile, the naturalness of an image block can be measured similarly under the GMM, which is learned on natural, original image patches. Therefore, general-purpose image forensics with good performance of fine-grained tampering localization is thereafter achieved by comparison of the two average patch log-likelihood values. The main advantages of the proposed method over the stateof-the-art image forensics are summarized in the following: The forensic methodology of the proposed method is very different from, and conceptually much simpler than the state-of-the-art image forensic methods. It does not need to extract hand-crafted features as almost all current forensic methods do. The proposed forensic method is general-purpose, which means it is able to cope with different image operations under the same likelihood comparison framework. It is also easy to be extended when a new image operation is taken into account. The single necessary step is to learn another GMM on image patches, processed by the considered new image operation. Since the image statistical models are constructed on small image patches, calculating the average patch loglikelihood value on a small image block is easy. Then, it is straightforward to perform fine-grained image tampering localization using the proposed method. The remainder of the paper is organized as follows. Sec. II presents the proposed general-purpose image forensic framework, with analysis of the suitability of the GMM for image forensic purposes. Experimental results are reported in Sec. III, with comparisons to a steganalytic feature SPAM [5], on both whole images and small image blocks for fine-grained tampering localization. Finally, we draw the conclusion in Sec. IV. A. Motivation II. PROPOSED METHOD As discussed in Sec. I, we hope to devise an image forensic method which is able to perform fine-grained tampering localization. Therefore, rather than analyzing a whole image, we are motivated to work on small image patches. In this paper, we consider the binary classification problem 2 whether a given image block has been processed. To this end, we aspire to find a way to measure the naturalness of small image patches. Between the two options natural and processed, if we are able to tell which one is the more likely case for a given image patch, then the classification is straightforward. To do so, it would be ideal if we are able to find a good distribution for natural, unprocessed patches, and that for the processed ones. As known, the distribution of natural images is very complicated. In the research field of natural image statistics, various attempts have appeared in order to find a good natural image statistical model in the spatial, transformed or filtered domain [14], which is the core of various computer vision problems as well as of image forensics and anti-forensics [15], [16]. Zoran and Weiss [17] firstly proposed to use the GMM for natural image patches. Serving as an excellent image prior model, it is proven to perform very well for various low-level computer vision problems such as denoising, deblurring, inpainting [17], and for JPEG image enhancement [18]. Compared with low-level computer vision problems where the GMM performs excellently as an image prior, image forensics is a very different problem. We are curious about whether the GMM is able to capture the statistical difference between natural, unprocessed images and the processed ones, that the human naked eyes fail to distinguish. To this end, we analyze the GMMs (see Sec. II-B for more details) with 200 components learned on 8 8 image patches extracted from the original images, JPEG images with quality factor 90, and sharpened images with filter parameter 0.5 (see Table I), respectively. If the GMM covariance matrices are able to successfully capture very different image statistics for different image operations, then the GMM will be a good choice for our image forensic task. For the above mentioned 3 GMMs, the eigenvectors of covariance matrices with size of the 4 leading components with the biggest mixing weights are shown in Fig. 1. The first GMM components mainly capture the contrast variability of image patches, whereas the following components reveal the textures and boundaries of objects [19]. However, even in the first 4 components, different patterns can be observed from different GMMs. Compared with Fig. 1-(a)-(d) obtained from the original images, we can see that JPEG compression is well exposed in (e)-(h), where the horizontal and vertical lines shown in the eigenvectors with relatively small eigenvalues actually depict the blocking artifacts present in the JPEG image. Compared with the JPEG images, the eigenvectors obtained from the sharpened images, well reflecting the detail enhancement effect, are even more different from those of the original images. 2 The image forensic work presented in this paper only considers binary classification, e.g., discriminating between the original and processed image blocks. The generality of image forensics is however embodied by the ability to cope with multiple image operations under the same forensic framework, though respectively considering different binary classification problems. Currently, we do not consider multi-class classification, which we plan to work on in the future.
3 π 1 = π 2 = π 3 = π 4 = ORI (a) (b) (c) (d) π 1 = π 2 = π 3 = π 4 = JPG (e) (f) (g) (h) π 1 = π 2 = π 3 = π 4 = USM (i) (j) Fig. 1. Eigenvectors of covariance matrices of the first 4 GMM components with the biggest mixing weights. In each subfigure, from the top left to the bottom right, the eigenvectors are sorted according to the descending order of the corresponding eigenvalues. (a)-(d) are obtained from the original images; (e)-(h) are obtained from the JPEG compressed images; and (i)-(l) are obtained from the processed images using the unsharp masking. The GMMs are learned on image dataset GFTR (see Sec. III-A). Different patterns appear in eigenvectors of covariance matrices of GMMs learned on different kinds of images. (k) (l) Based on the above natural image statistics investigation and experimental analysis of different GMMs learned on different kinds of images, we adopt the GMM as the image statistical model in the proposed forensic method. Under this model, we are able to measure the naturalness of a given image patch by likelihood calculation. By further average patch loglikelihood comparison of a given image block, we can thereafter conduct forensic classification. Besides the simplicity and being very expressive for small image patches processed by different operations (as shown in Fig. 1), we also notice that the GMM may also bring generality to the proposed image forensic framework. This is because we can easily integrate a new image processing operation, as long as we are able to correspondingly learn a distinctive GMM. B. Forensic Analysis Using Image Statistical Models Based on the discussion and analysis in Sec. II-A, we adopt the GMM to model the distribution of small image patches. Given a generic image patch x (in vectorized form, its original size is b b before stacking), its likelihood under a GMM is computed by: L(θ x) = p(x θ) = K π k N(x µ k,c k ), (1) k=1 where π k, µ k and C k are respectively the mixing weight, mean and covariance matrix for the k-th (k = 1,, K) GMM component. For sake of brevity, θ = {π k,µ k,c k k = 1,,K} denotes the parameters describing the GMM. The calculated likelihood value L(θ x) indicates how likely x follows the GMM distribution parametrized by θ. From the patches extracted from images processed by a certain image operation (or no processing at all), the parameters θ of the GMM can be learned using the standard EM (Expectation Maximization) algorithm 3. Therefore, we build a parametric image statistical model for image patches corresponding to each kind of image operation in consideration. For the binary classification of our image forensic problem, we are given a generic image (block) X with size 3 In practice, we use the unoptimized Matlab code which can be downloaded from: em-algorithm-for-gaussian-mixture-model.
4 H W (H,W b), and we are to give a binary decision about its processing history. From X, we can extract a set of N = (H b+1) (W b+1) overlapping image patches of size b b: {x i i = 1,,N}. Therefore, the average patch log-likelihood (also called expected patch log-likelihood in [17]) of X can be calculated as 1 N N i=1 log L(θ x i). We assume that X is original under the null hypothesis H 0, and is processed by a certain image operation under the alternative hypothesis H 1. Under H 0, the patches are assumed to follow the GMM parametrized by θ 0 ; while under H 1, the patches are assumed to follow the GMM parametrized by θ 1, whose value varies across different image operations. Therefore, the binary classification problem of whether X has been processed, can be formulated as a simple hypothesis testing problem, and we propose the following test on the difference of two average patch log-likelihood values: Λ(X) = 1 N log L(θ 0 x i ) 1 N log L(θ 1 x i ) η. (2) N N i=1 i=1 The decision rule of the test is as follows: { reject H0 if Λ(X) η do not reject H 0 if Λ(X) > η. For image forensic classification, if Λ(X) η, we classify the image block X as processed, otherwise as original. A. Experimental Setting III. EXPERIMENTAL RESULTS In this paper, we consider 6 different image operations 4 as summarized in Table I and listed below: Gaussian filtering is often used for image smoothing, in order to remove noise or to reduce details; JPEG is one of the most widely used image compression formats today, which is a popular choice of image forgers; Median filtering is a commonly used image smoothing technique, which is particularly effective for removing impulsive noise. It can also be used to hide artifacts of JPEG compression [20] and resampling [21] 5 ; Resampling is often involved in creating composite image forgeries, where the size or angle of one source image needs to be adjusted; Unsharp masking is a popular image sharpening technique, to create less blurry, enhanced image than the original; White Gaussian noise addition, despite of its rare use in conventional image processing, yet one can find its applications in disguising traces of other image operations, e.g., for JPEG deblocking purposes [20]. For the sake of brevity, the abbreviations in Table I are used to refer to different image operations. 4 Currently, we only consider one parameter setting for each image operation. We will consider multiple parameter settings and include more kinds of image operations in the follow-up work. 5 Indeed, we consider median filtering or to-be-described white Gaussian noise addition as an image operation motivated by certain image anti-forensic methods. Currently, we however do not test against anti-forensics, which is left as a future effort. (3) ORI GF TABLE I DIFFERENT IMAGE OPERATIONS CONSIDERED IN THIS PAPER. no image processing Gaussian filtering with window size 3 3, and standard deviation 0.5 to generate the filter kernel JPG JPEG compression with quality factor 90 MF median filtering with window size 3 3 RS USM resampling with bicubic interpolation to scale the image to 80% of its original size unsharp masking with window size 3 3, and parameter 0.5 for the Laplacian filter to generate the sharpening filter kernel WGN white Gaussian noise addition with standard deviation 2 The natural image datasets used in this paper are created from 545 never resampled, non-compressed TIFF images 6, with various indoor and outdoor scenes. They were taken by 4 cameras of different makes and models, and have been used for image resampling forensics [3] and double JPEG compression forensics [22]. These images are randomly divided into two sets, which respectively include 273 images for training and 272 images for testing. From each original high-resolution TIFF image, we crop 9 adjacent subimages of size from its center. Without loss of generality, we only consider grayscale images in this paper. Therefore, we convert the cropped TIFF images into 8-bit grayscale images using the Matlab function rgb2gray. In the end, we have = 2457 images in the training dataset GFTR (Generalpurpose Forensic TRaining), and = 2448 images in the testing dataset GFTE (General-purpose Forensic TEsting). In order to evaluate the forensic performance of the proposed method on exposing image processing operation on whole images as well as on small image blocks, we consider 3 image (block) sizes H W : , 32 32, and B. Forensic Performance For experimental comparison, we construct forensic detectors based on the well-known 686-dimensional 2nd-order SPAM feature [5], which was initially designed for steganalysis. Its effectiveness in both steganalysis and general-purpose image forensics has been well demonstrated in literature [4], [8]. Moreover, we choose to compare with the steganalytic feature SPAM instead of the SRM (34671-dimensional) [6] and LBP (22153-dimensional) [7] features, partly because of its relatively low dimensionality. We are interested in finegrained tampering localization and the performance of forensic detectors on small image blocks (down to block size with only 256 pixels in our experimental setting). Not only are the feature extraction and detector training of SRM and LBP computationally demanding, but the obtained features from small image blocks may also contain lots of redundancies. For each image (block) size H W and each image operation, a SPAM-based detector is trained using the SVM [12] on the original and the corresponding processed image(s) 6 Downloaded from: ftp://firewall.teleco.uvigo.es:27244/ds 01 UTFI.zip and ftp://lesc.dinfo.unifi.it/pub/public/jpegloc/dataset/.
5 (a) (b) (f) (c) (g) (d) (e) (h) (i) Fig. 2. Example results of fine-grained image tampering localization to reveal forgery semantics, with block size Forgery (c) is created by splicing the original image (a) and the church image area extracted from the JPEG image (b) with quality factor 90. (d) points out the JPEG compressed image blocks using the proposed method, and (e) reports the results obtained by the SPAM-based detector [5]. Forgery (g) is created by scaling the dog image area of the original image (f) to 80% of its original size and insert it back to the same image (f). (h) shows the image blocks detected as resampled using the proposed method, whereas (i) presents the results obtained using the SPAM-based detector [5]. Images (a), (f), and the corresponding non-compressed image of (b) are from the GFTE dataset. Results obtained on these two image forgeries indicate the proposed method achieves a better forensic performance on fine-grained tampering localization, because of its high detection accuracy on small image blocks. (blocks) from the GFTR dataset7. For learning the GMMs for the proposed method, we consider a patch size of 8 8 (namely b = 8), which well balances the richness of the model and the complexity of the learning. We perform the learning procedure on million 8 8 image patches extracted from dataset GFTR, with pixel values scaled to [0, 1] and DC component removed. In the end, we obtain 200 components for each GMM. We have two strategies to set the threshold η in Eq. (3), in order to build the proposed forensic detectors. The straightforward setting is η = 0, meaning that X is classified as processed when the test measurement Λ(X) indicates that its extracted patches are on average more likely to be under the GMM of the processed image patches. The more sophisticated setting of η can be implemented according to the detector s performance on the GFTR dataset, by maximizing the detection accuracy. For forensic testing, we use the 2448 images in GFTE dataset for the size ; whereas for the sizes and 16 16, 10 blocks are randomly sampled from each (processed) GFTE image, resulting in blocks corresponding to each kind of image operation. Table II reports the detection accuracy for different binary classification problems between the original and the processed image(s) (blocks). Here, SPAM-based indicates 7 For image block size and 16 16, one block is randomly sampled from each GFTR image, for training the SPAM-based detectors. In practice, they have been proven to perform slightly better (around 2% detection accuracy improvement) on image blocks with the corresponding size than SPAM-based detectors trained on images. TABLE II C LASSIFICATION ACCURACY (%) COMPARISON WHEN TESTED ON DIFFERENT SIZES OF IMAGE ( BLOCK ). T HE SPAM- BASED DETECTORS AND THE GMM S ARE TRAINED ON DATASET GFTR. R ESULTS ARE OBTAINED BY TESTS ON DATASET GFTE. SPAM-based Proposed-S Proposed-T SPAM-based Proposed-S Proposed-T SPAM-based Proposed-S Proposed-T GF JPG MF RS USM WGN the results achieved by the SPAM-based detectors [5]; whereas Proposed-S and Proposed-T show results achieved by the proposed detectors by using η = 0 and the trained η values from the GFTR dataset, respectively. In order to save some computation cost for the proposed method, for the image, not all the overlapping b b patches are taken into account. Instead, the non-overlapping patches of the image and those of the corresponding image cropped by 4 pixels horizontally and vertically are used. In practice, this simplification strategy does not harm the forensic performance of the proposed method, and can largely reduce the computation cost. We can see from Table II, the proposed detectors are able to achieve comparable performance with
6 respect to the SPAM-based ones, when tested on images. The proposed method is especially advantageous when tested on small image blocks, especially for the size In particular, for resampling, the proposed method obtains 11.5% gain of detection accuracy over the SPAM feature. Even though the proposed method does not outperform the SPAM feature for median filtering, its detection accuracy is still around 93%. Moreover, the forensic performance of the SPAM feature drops, especially for JPG and RS, when the size of testing image blocks decreases, while ours stays rather stable for all the processing operations in consideration. The good performance of the proposed method for exposing image processing operation on small image blocks demonstrates its potential for fine-grained image tampering localization. Besides the large-scale test results shown in Table II, here we test on two image forgeries where JPEG compression and resampling are involved. As shown in Fig. 2- (c), it is a composite image forgery by inserting the church image area of the JPEG image shown in -(b) into the original image -(a). The small dog on the upper right of Fig. 2- (g) is actually a copy of the dog image area on the bottom left after resizing it. The two forgeries are divided into nonoverlapping blocks and are thereafter fed to the forensic detectors for block-wise binary classification. The small red boxes in Fig. 2-(d) and -(h) show the detection results using the proposed method for JPEG compression and resampling, respectively. Compared with results shown in Fig. 2-(e) and -(i) obtained by the SPAM-based detectors [5], we can see that our method successfully reveals, with a higher accuracy, the tampered image areas at a fine-grained level. IV. CONCLUSION Driven by the question whether a given image block is more like a natural, original block or a processed one, this paper adopts the GMM to model the statistics of images processed by different image operations. Given an image block, by comparing the average patch log-likelihood values calculated under two GMMs learned from the original and processed images, we are able to perform binary classification for forensic purposes. Very different from the existing image forensic methods, the proposed method is conceptually simple in methodology, does not need complex feature extraction nor the adoption of the ensemble classifier or the SVM, is easy to be extended for new image operations, and is able to perform fine-grained tampering localization. It is especially advantageous in tracing image processing history on small image blocks, and is able to achieve comparable performance on images, compared with the 2nd-order SPAM feature [5]. Future research shall be devoted to investigation of richer image statistical models, multi-class forensic classification, integration of more image operations as well as more of their parameter settings, comparison with detectors based on steganalytic features SRM [6] and LBP [7], and tests against image anti-forensic methods. ACKNOWLEDGMENT This work is supported, in part by the AGIR (Alpes Grenoble Innovation Recherche) project TAFFI, in part by the French Rhône- Alpes region through the CMIRA Scholarship Program, in part by the French Eiffel Scholarship under Grant B, and in part by the French ANR Estampille under Grant ANR-10-CORD-019. REFERENCES [1] W. Luo, J. Huang, and G. Qiu, JPEG error analysis and its applications to digital image forensics, IEEE Trans. Inf. Forensics Security, vol. 5, no. 3, pp , [2] X. Kang, M. C. Stamm, A. Peng, and K. J. R. Liu, Robust median filtering forensics using an autoregressive model, IEEE Trans. Inf. Forensics Security, vol. 8, no. 9, pp , [3] D. Vázquez-Padín and F. Pérez-González, ML estimation of the resampling factor, in Proc. IEEE Int. Workshop on Information Forensics and Security, 2012, pp [4] X. Qiu, H. Li, W. Luo, and J. Huang, A universal image forensic strategy based on steganalytic model, in Proc. ACM Int. Workshop on Information Hiding and Multimedia Security, 2014, pp [5] T. Pevný, P. Bas, and J. Fridrich, Steganalysis by subtractive pixel adjacency matrix, IEEE Trans. Inf. Forensics Security, vol. 5, no. 2, pp , [6] J. Fridrich and J. Kodovský, Rich models for steganalysis of digital images, IEEE Trans. Inf. Forensics Security, vol. 7, no. 3, pp , [7] Y. Q. Shi, P. Sutthiwan, and L. Chen, Textural features for steganalysis, in Proc. Int. Conf. on Information Hiding, 2012, pp [8] M. Kirchner and J. Fridrich, On detection of median filtering in digital images, in Proc. IS&T/SPIE Electronic Imaging, vol. 7541, 2010, p [9] H. Li, W. Luo, and J. Huang, Countering anti-jpeg compression forensics, in Proc. IEEE Int. Conf. Image Process., 2012, pp [10] C. Chen and Y. Q. Shi, JPEG image steganalysis utilizing both intrablock and interblock correlations, in Proc. IEEE Int. Symp. on Circuits Syst., 2008, pp [11] J. Kodovský, J. Fridrich, and V. Holub, Ensemble classifiers for steganalysis of digital media, IEEE Trans. Inf. Forensics Security, vol. 7, no. 2, pp , [12] C.-C. Chang and C.-J. Lin, LIBSVM: A library for support vector machines, ACM Trans. Intell. Syst. Technol., vol. 2, no. 3, 2011, Art. ID 27. [13] G. J. Simmons, The prisoners problem and the subliminal channel, in Advances in Cryptology: Proc. of CRYPTO, 1983, pp [14] E. P. Simoncelli and B. A. Olshausen, Natural image statistics and neural representation, Annu. Rev. Neurosci., vol. 24, pp , [15] R. Böhme and M. Kirchner, Counter-forensics: Attacking image forensics, in Digital Image Forensics, H. T. Sencar and N. Memon, Eds. New York, NY, USA: Springer, 2013, pp [16] W. Fan, K. Wang, F. Cayre, and Z. Xiong, Median filtered image quality enhancement and anti-forensics via variational deconvolution, IEEE Trans. Inf. Forensics Security, vol. 10, no. 5, pp , [17] D. Zoran and Y. Weiss, From learning models of natural image patches to whole image restoration, in Proc. IEEE Int. Conf. Comput. Vis., 2011, pp [18] W. Fan, K. Wang, F. Cayre, and Z. Xiong, JPEG anti-forensics using non-parametric DCT quantization noise estimation and natural image statistics, in Proc. ACM Int. Workshop on Information Hiding and Multimedia Security, 2013, pp [19] D. Zoran and Y. Weiss, Natural images, Gaussian mixtures and dead leaves, in Adv. Neural Inf. Process. Syst., 2012, pp [20] M. C. Stamm and K. J. R. Liu, Anti-forensics of digital image compression, IEEE Trans. Inf. Forensics Security, vol. 6, no. 3, pp , [21] M. Kirchner and R. Röhme, Hiding traces of resampling in digital images, IEEE Trans. Inf. Forensics Security, vol. 3, no. 4, pp , [22] T. Bianchi and A. Piva, Image forgery localization via block-grained analysis of JPEG artifacts, IEEE Trans. Inf. Forensics Security, vol. 7, no. 3, pp , 2012.
Detecting Resized Double JPEG Compressed Images Using Support Vector Machine
Detecting Resized Double JPEG Compressed Images Using Support Vector Machine Hieu Cuong Nguyen and Stefan Katzenbeisser Computer Science Department, Darmstadt University of Technology, Germany {cuong,katzenbeisser}@seceng.informatik.tu-darmstadt.de
More informationIMPROVEMENTS ON SOURCE CAMERA-MODEL IDENTIFICATION BASED ON CFA INTERPOLATION
IMPROVEMENTS ON SOURCE CAMERA-MODEL IDENTIFICATION BASED ON CFA INTERPOLATION Sevinc Bayram a, Husrev T. Sencar b, Nasir Memon b E-mail: sevincbayram@hotmail.com, taha@isis.poly.edu, memon@poly.edu a Dept.
More informationImage Tampering Localization via Estimating the Non-Aligned Double JPEG compression
Image Tampering Localization via Estimating the Non-Aligned Double JPEG compression Lanying Wu a, Xiangwei Kong* a, Bo Wang a, Shize Shang a a School of Information and Communication Engineering, Dalian
More informationIMAGE TAMPERING DETECTION BY EXPOSING BLUR TYPE INCONSISTENCY. Khosro Bahrami and Alex C. Kot
24 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) IMAGE TAMPERING DETECTION BY EXPOSING BLUR TYPE INCONSISTENCY Khosro Bahrami and Alex C. Kot School of Electrical and
More informationA Simple and Effective Image-Statistics-Based Approach to Detecting Recaptured Images from LCD Screens
A Simple and Effective Image-Statistics-Based Approach to Detecting Recaptured Images from LCD Screens Kai Wang Univ. Grenoble Alpes, CNRS, Grenoble INP, GIPSA-lab, 38000 Grenoble, France Abstract It is
More informationIntroduction to Video Forgery Detection: Part I
Introduction to Video Forgery Detection: Part I Detecting Forgery From Static-Scene Video Based on Inconsistency in Noise Level Functions IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, VOL. 5,
More informationIDENTIFYING DIGITAL CAMERAS USING CFA INTERPOLATION
Chapter 23 IDENTIFYING DIGITAL CAMERAS USING CFA INTERPOLATION Sevinc Bayram, Husrev Sencar and Nasir Memon Abstract In an earlier work [4], we proposed a technique for identifying digital camera models
More informationImage Manipulation Detection using Convolutional Neural Network
Image Manipulation Detection using Convolutional Neural Network Dong-Hyun Kim 1 and Hae-Yeoun Lee 2,* 1 Graduate Student, 2 PhD, Professor 1,2 Department of Computer Software Engineering, Kumoh National
More informationINFORMATION about image authenticity can be used in
1 Constrained Convolutional Neural Networs: A New Approach Towards General Purpose Image Manipulation Detection Belhassen Bayar, Student Member, IEEE, and Matthew C. Stamm, Member, IEEE Abstract Identifying
More informationDetection of Misaligned Cropping and Recompression with the Same Quantization Matrix and Relevant Forgery
Detection of Misaligned Cropping and Recompression with the Same Quantization Matrix and Relevant Forgery Qingzhong Liu Department of Computer Science Sam Houston State University Huntsville, TX 77341,
More informationPRIOR IMAGE JPEG-COMPRESSION DETECTION
Applied Computer Science, vol. 12, no. 3, pp. 17 28 Submitted: 2016-07-27 Revised: 2016-09-05 Accepted: 2016-09-09 Compression detection, Image quality, JPEG Grzegorz KOZIEL * PRIOR IMAGE JPEG-COMPRESSION
More informationLiterature Survey on Image Manipulation Detection
Literature Survey on Image Manipulation Detection Rani Mariya Joseph 1, Chithra A.S. 2 1M.Tech Student, Computer Science and Engineering, LMCST, Kerala, India 2 Asso. Professor, Computer Science And Engineering,
More informationISSN: ISO 9001:2008 Certified International Journal of Engineering and Innovative Technology (IJEIT) Volume 2, Issue 3, September 2012
A Tailored Anti-Forensic Approach for Digital Image Compression S.Manimurugan, Athira B.Kaimal Abstract- The influence of digital images on modern society is incredible; image processing has now become
More informationHigher-Order, Adversary-Aware, Double JPEG-Detection via Selected Training on Attacked Samples
Higher-Order, Adversary-Aware, Double JPEG-Detection via Selected Training on ed Samples Mauro Barni, Ehsan Nowroozi, Benedetta Tondi Department of Information Engineering and Mathematics, University of
More informationCountering Anti-Forensics of Lateral Chromatic Aberration
IH&MMSec 7, June -, 7, Philadelphia, PA, USA Countering Anti-Forensics of Lateral Chromatic Aberration Owen Mayer Drexel University Department of Electrical and Computer Engineering Philadelphia, PA, USA
More informationDetection of Image Forgery was Created from Bitmap and JPEG Images using Quantization Table
Detection of Image Forgery was Created from Bitmap and JPEG Images using Quantization Tran Dang Hien University of Engineering and Eechnology, VietNam National Univerity, VietNam Pham Van At Department
More informationReversible Data Hiding in JPEG Images Based on Adjustable Padding
Reversible Data Hiding in JPEG Images Based on Adjustable Padding Ching-Chun Chang Department of Computer Science University of Warwick United Kingdom Email: C.Chang.@warwick.ac.uk Chang-Tsun Li School
More informationApplication of Histogram Examination for Image Steganography
J. Appl. Environ. Biol. Sci., 5(9S)97-104, 2015 2015, TextRoad Publication ISSN: 2090-4274 Journal of Applied Environmental and Biological Sciences www.textroad.com Application of Histogram Examination
More informationIJSRD - International Journal for Scientific Research & Development Vol. 4, Issue 01, 2016 ISSN (online):
IJSRD - International Journal for Scientific Research & Development Vol. 4, Issue 01, 2016 ISSN (online): 2321-0613 High-Quality Jpeg Compression using LDN Comparison and Quantization Noise Analysis S.Sasikumar
More informationSteganalysis in resized images
Steganalysis in resized images Jan Kodovský, Jessica Fridrich ICASSP 2013 1 / 13 Outline 1. Steganography basic concepts 2. Why we study steganalysis in resized images 3. Eye-opening experiment on BOSSbase
More informationImage Forgery Detection Using Svm Classifier
Image Forgery Detection Using Svm Classifier Anita Sahani 1, K.Srilatha 2 M.E. Student [Embedded System], Dept. Of E.C.E., Sathyabama University, Chennai, India 1 Assistant Professor, Dept. Of E.C.E, Sathyabama
More informationSteganalytic methods for the detection of histogram shifting data-hiding schemes
Steganalytic methods for the detection of histogram shifting data-hiding schemes Daniel Lerch and David Megías Universitat Oberta de Catalunya, Spain. ABSTRACT In this paper, some steganalytic techniques
More informationCamera Model Identification Framework Using An Ensemble of Demosaicing Features
Camera Model Identification Framework Using An Ensemble of Demosaicing Features Chen Chen Department of Electrical and Computer Engineering Drexel University Philadelphia, PA 19104 Email: chen.chen3359@drexel.edu
More informationDr. Kusam Sharma *1, Prof. Pawanesh Abrol 2, Prof. Devanand 3 ABSTRACT I. INTRODUCTION
International Journal of Scientific Research in Computer Science, Engineering and Information Technology 2017 IJSRCSEIT Volume 2 Issue 6 ISSN : 2456-3307 Feature Based Analysis of Copy-Paste Image Tampering
More informationDetection of Adaptive Histogram Equalization Robust Against JPEG Compression
Detection of Adaptive Histogram Equalization Robust Against JPEG Compression Mauro Barni, Ehsan Nowroozi, Benedetta Tondi Department of Information Engineering and Mathematics, University of Siena Via
More informationSource Camera Model Identification Using Features from contaminated Sensor Noise
Source Camera Model Identification Using Features from contaminated Sensor Noise Amel TUAMA 2,3, Frederic COMBY 2,3, Marc CHAUMONT 1,2,3 1 NÎMES UNIVERSITY, F-30021 Nîmes Cedex 1, France 2 MONTPELLIER
More informationEFFECT OF SATURATED PIXELS ON SECURITY OF STEGANOGRAPHIC SCHEMES FOR DIGITAL IMAGES. Vahid Sedighi and Jessica Fridrich
EFFECT OF SATURATED PIXELS ON SECURITY OF STEGANOGRAPHIC SCHEMES FOR DIGITAL IMAGES Vahid Sedighi and Jessica Fridrich Binghamton University Department of ECE Binghamton, NY ABSTRACT When hiding messages
More informationImproved Detection of LSB Steganography in Grayscale Images
Improved Detection of LSB Steganography in Grayscale Images Andrew Ker adk@comlab.ox.ac.uk Royal Society University Research Fellow at Oxford University Computing Laboratory Information Hiding Workshop
More informationLocal prediction based reversible watermarking framework for digital videos
Local prediction based reversible watermarking framework for digital videos J.Priyanka (M.tech.) 1 K.Chaintanya (Asst.proff,M.tech(Ph.D)) 2 M.Tech, Computer science and engineering, Acharya Nagarjuna University,
More informationA Deep Learning Approach To Universal Image Manipulation Detection Using A New Convolutional Layer
A Deep Learning Approach To Universal Image Manipulation Detection Using A New Convolutional Layer ABSTRACT Belhassen Bayar Drexel University Dept. of ECE Philadelphia, PA, USA bb632@drexel.edu When creating
More informationAn Implementation of LSB Steganography Using DWT Technique
An Implementation of LSB Steganography Using DWT Technique G. Raj Kumar, M. Maruthi Prasada Reddy, T. Lalith Kumar Electronics & Communication Engineering #,JNTU A University Electronics & Communication
More informationForgery Detection using Noise Inconsistency: A Review
Forgery Detection using Noise Inconsistency: A Review Savita Walia, Mandeep Kaur UIET, Panjab University Chandigarh ABSTRACT: The effects of digital forgeries and image manipulations may not be seen by
More informationDigital Image Sharing and Removing the Transmission Risk Problem by Using the Diverse Image Media
1 1 Digital Image Sharing and Removing the Transmission Risk Problem by Using the Diverse Image Media 1 Shradha S. Rathod, 2 Dr. D. V. Jadhav, 1 PG Student, 2 Principal, 1,2 TSSM s Bhivrabai Sawant College
More informationForensic Hash for Multimedia Information
Forensic Hash for Multimedia Information Wenjun Lu, Avinash L. Varna and Min Wu Department of Electrical and Computer Engineering, University of Maryland, College Park, U.S.A email: {wenjunlu, varna, minwu}@eng.umd.edu
More informationCamera identification from sensor fingerprints: why noise matters
Camera identification from sensor fingerprints: why noise matters PS Multimedia Security 2010/2011 Yvonne Höller Peter Palfrader Department of Computer Science University of Salzburg January 2011 / PS
More informationGlobal Contrast Enhancement Detection via Deep Multi-Path Network
Global Contrast Enhancement Detection via Deep Multi-Path Network Cong Zhang, Dawei Du, Lipeng Ke, Honggang Qi School of Computer and Control Engineering University of Chinese Academy of Sciences, Beijing,
More informationDYNAMIC CONVOLUTIONAL NEURAL NETWORK FOR IMAGE SUPER- RESOLUTION
Journal of Advanced College of Engineering and Management, Vol. 3, 2017 DYNAMIC CONVOLUTIONAL NEURAL NETWORK FOR IMAGE SUPER- RESOLUTION Anil Bhujel 1, Dibakar Raj Pant 2 1 Ministry of Information and
More informationWavelet-based Image Splicing Forgery Detection
Wavelet-based Image Splicing Forgery Detection 1 Tulsi Thakur M.Tech (CSE) Student, Department of Computer Technology, basiltulsi@gmail.com 2 Dr. Kavita Singh Head & Associate Professor, Department of
More informationExposing Digital Forgeries from JPEG Ghosts
1 Exposing Digital Forgeries from JPEG Ghosts Hany Farid, Member, IEEE Abstract When creating a digital forgery, it is often necessary to combine several images, for example, when compositing one person
More informationRetrieval of Large Scale Images and Camera Identification via Random Projections
Retrieval of Large Scale Images and Camera Identification via Random Projections Renuka S. Deshpande ME Student, Department of Computer Science Engineering, G H Raisoni Institute of Engineering and Management
More informationThe Influence of Image Enhancement Filters on a Watermark Detection Rate Authors
acta graphica 194 udc 004.056.55:655.36 original scientific paper received: -09-011 accepted: 11-11-011 The Influence of Image Enhancement Filters on a Watermark Detection Rate Authors Ante Poljičak, Lidija
More informationCorrelation Based Image Tampering Detection
Correlation Based Image Tampering Detection Priya Singh M. Tech. Scholar CSE Dept. MIET Meerut, India Abstract-The current era of digitization has made it easy to manipulate the contents of an image. Easy
More informationCS 365 Project Report Digital Image Forensics. Abhijit Sharang (10007) Pankaj Jindal (Y9399) Advisor: Prof. Amitabha Mukherjee
CS 365 Project Report Digital Image Forensics Abhijit Sharang (10007) Pankaj Jindal (Y9399) Advisor: Prof. Amitabha Mukherjee 1 Abstract Determining the authenticity of an image is now an important area
More informationMultimedia Forensics
Multimedia Forensics Using Mathematics and Machine Learning to Determine an Image's Source and Authenticity Matthew C. Stamm Multimedia & Information Security Lab (MISL) Department of Electrical and Computer
More informationTHE popularization of imaging components equipped in
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, VOL. 10, NO. 3, MARCH 2015 Revealing the Trace of High-Quality JPEG Compression Through Quantization Noise Analysis Bin Li, Member, IEEE, Tian-Tsong
More informationA New Scheme for No Reference Image Quality Assessment
Author manuscript, published in "3rd International Conference on Image Processing Theory, Tools and Applications, Istanbul : Turkey (2012)" A New Scheme for No Reference Image Quality Assessment Aladine
More informationMoving Object Detection for Intelligent Visual Surveillance
Moving Object Detection for Intelligent Visual Surveillance Ph.D. Candidate: Jae Kyu Suhr Advisor : Prof. Jaihie Kim April 29, 2011 Contents 1 Motivation & Contributions 2 Background Compensation for PTZ
More informationLicense Plate Localisation based on Morphological Operations
License Plate Localisation based on Morphological Operations Xiaojun Zhai, Faycal Benssali and Soodamani Ramalingam School of Engineering & Technology University of Hertfordshire, UH Hatfield, UK Abstract
More informationWITH the rapid development of image processing technology,
480 IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, VOL. 5, NO. 3, SEPTEMBER 2010 JPEG Error Analysis and Its Applications to Digital Image Forensics Weiqi Luo, Member, IEEE, Jiwu Huang, Senior
More informationSteganalysis of Overlapping Images
Steganalysis of Overlapping Images Jimmy Whitaker JimmyMWhitaker @ gmail.com Andrew Ker adk@ cs.ox.ac.uk SPIE/IS&T Electronic Imaging, San Francisco, 11 February 2015 Real-world images Real-world images
More informationISSN (PRINT): , (ONLINE): , VOLUME-4, ISSUE-11,
FPGA IMPLEMENTATION OF LSB REPLACEMENT STEGANOGRAPHY USING DWT M.Sathya 1, S.Chitra 2 Assistant Professor, Prince Dr. K.Vasudevan College of Engineering and Technology ABSTRACT An enhancement of data protection
More informationDigital Image Processing 3/e
Laboratory Projects for Digital Image Processing 3/e by Gonzalez and Woods 2008 Prentice Hall Upper Saddle River, NJ 07458 USA www.imageprocessingplace.com The following sample laboratory projects are
More informationFragile Sensor Fingerprint Camera Identification
Fragile Sensor Fingerprint Camera Identification Erwin Quiring Matthias Kirchner Binghamton University IEEE International Workshop on Information Forensics and Security Rome, Italy November 19, 2015 Camera
More informationHistogram Modification Based Reversible Data Hiding Using Neighbouring Pixel Differences
Histogram Modification Based Reversible Data Hiding Using Neighbouring Pixel Differences Ankita Meenpal*, Shital S Mali. Department of Elex. & Telecomm. RAIT, Nerul, Navi Mumbai, Mumbai, University, India
More informationLocating Steganographic Payload via WS Residuals
Locating Steganographic Payload via WS Residuals Andrew D. Ker Oxford University Computing Laboratory Parks Road Oxford OX1 3QD, UK adk@comlab.ox.ac.uk ABSTRACT The literature now contains a number of
More informationAn Efficient Noise Removing Technique Using Mdbut Filter in Images
IOSR Journal of Electronics and Communication Engineering (IOSR-JECE) e-issn: 2278-2834,p- ISSN: 2278-8735.Volume 10, Issue 3, Ver. II (May - Jun.2015), PP 49-56 www.iosrjournals.org An Efficient Noise
More informationExposing Image Forgery with Blind Noise Estimation
Exposing Image Forgery with Blind Noise Estimation Xunyu Pan Computer Science Department University at Albany, SUNY Albany, NY 12222, USA xypan@cs.albany.edu Xing Zhang Computer Science Department University
More informationSOURCE CAMERA IDENTIFICATION BASED ON SENSOR DUST CHARACTERISTICS
SOURCE CAMERA IDENTIFICATION BASED ON SENSOR DUST CHARACTERISTICS A. Emir Dirik Polytechnic University Department of Electrical and Computer Engineering Brooklyn, NY, US Husrev T. Sencar, Nasir Memon Polytechnic
More informationCopy-Move Image Forgery Detection using SVD
Copy-Move Image Forgery Detection using SVD Mr. Soumen K. Patra 1, Mr. Abhijit D. Bijwe 2 1M. Tech in Communication, Department of Electronics & Communication, Priyadarshini Institute of Engineering &
More informationAUGMENTED CONVOLUTIONAL FEATURE MAPS FOR ROBUST CNN-BASED CAMERA MODEL IDENTIFICATION. Belhassen Bayar and Matthew C. Stamm
AUGMENTED CONVOLUTIONAL FEATURE MAPS FOR ROBUST CNN-BASED CAMERA MODEL IDENTIFICATION Belhassen Bayar and Matthew C. Stamm Department of Electrical and Computer Engineering, Drexel University, Philadelphia,
More informationJournal of mathematics and computer science 11 (2014),
Journal of mathematics and computer science 11 (2014), 137-146 Application of Unsharp Mask in Augmenting the Quality of Extracted Watermark in Spatial Domain Watermarking Saeed Amirgholipour 1 *,Ahmad
More informationImage Forgery Identification Using JPEG Intrinsic Fingerprints
1 Image Forgery Identification Using JPEG Intrinsic Fingerprints A. Garg, A. Hailu, and R. Sridharan Abstract In this paper a novel method for image forgery detection is presented. he method exploits the
More information2018 IEEE Signal Processing Cup: Forensic Camera Model Identification Challenge
2018 IEEE Signal Processing Cup: Forensic Camera Model Identification Challenge This competition is sponsored by the IEEE Signal Processing Society Introduction The IEEE Signal Processing Society s 2018
More informationDigital Watermarking Using Homogeneity in Image
Digital Watermarking Using Homogeneity in Image S. K. Mitra, M. K. Kundu, C. A. Murthy, B. B. Bhattacharya and T. Acharya Dhirubhai Ambani Institute of Information and Communication Technology Gandhinagar
More informationImage De-Noising Using a Fast Non-Local Averaging Algorithm
Image De-Noising Using a Fast Non-Local Averaging Algorithm RADU CIPRIAN BILCU 1, MARKKU VEHVILAINEN 2 1,2 Multimedia Technologies Laboratory, Nokia Research Center Visiokatu 1, FIN-33720, Tampere FINLAND
More informationIMAGE SPLICING FORGERY DETECTION AND LOCALIZATION USING FREQUENCY-BASED FEATURES
Chiew K.T., et al. (Eds.): PGRES 2017, Kuala Lumpur: Eastin Hotel, FCSIT, 2017: pp 35-42 IMAGE SPLICING FORGERY DETECTION AND LOCALIZATION USING FREQUENCY-BASED FEATURES Thamarai Subramaniam and Hamid
More informationExploration of Least Significant Bit Based Watermarking and Its Robustness against Salt and Pepper Noise
Exploration of Least Significant Bit Based Watermarking and Its Robustness against Salt and Pepper Noise Kamaldeep Joshi, Rajkumar Yadav, Sachin Allwadhi Abstract Image steganography is the best aspect
More informationRobust Multi-Classifier for Camera Model Identification Based on Convolution Neural Network
Received March 14, 2018, accepted April 20, 2018, date of publication May 1, 2018, date of current version May 24, 2018. Digital Object Identifier 10.1109/ACCESS.2018.2832066 Robust Multi-Classifier for
More informationWatermarking-based Image Authentication with Recovery Capability using Halftoning and IWT
Watermarking-based Image Authentication with Recovery Capability using Halftoning and IWT Luis Rosales-Roldan, Manuel Cedillo-Hernández, Mariko Nakano-Miyatake, Héctor Pérez-Meana Postgraduate Section,
More informationSTEGANOGRAPHY WITH TWO JPEGS OF THE SAME SCENE. Tomáš Denemark, Student Member, IEEE, and Jessica Fridrich, Fellow, IEEE
STEGANOGRAPHY WITH TWO JPEGS OF THE SAME SCENE Tomáš Denemark, Student Member, IEEE, and Jessica Fridrich, Fellow, IEEE Binghamton University Department of ECE Binghamton, NY ABSTRACT It is widely recognized
More informationDifferent-quality Re-demosaicing in Digital Image Forensics
Different-quality Re-demosaicing in Digital Image Forensics 1 Bo Wang, 2 Xiangwei Kong, 3 Lanying Wu *1,2,3 School of Information and Communication Engineering, Dalian University of Technology E-mail:
More informationThis content has been downloaded from IOPscience. Please scroll down to see the full text.
This content has been downloaded from IOPscience. Please scroll down to see the full text. Download details: IP Address: 148.251.232.83 This content was downloaded on 10/07/2018 at 03:39 Please note that
More informationMain Subject Detection of Image by Cropping Specific Sharp Area
Main Subject Detection of Image by Cropping Specific Sharp Area FOTIOS C. VAIOULIS 1, MARIOS S. POULOS 1, GEORGE D. BOKOS 1 and NIKOLAOS ALEXANDRIS 2 Department of Archives and Library Science Ionian University
More informationPRACTICAL IMAGE AND VIDEO PROCESSING USING MATLAB
PRACTICAL IMAGE AND VIDEO PROCESSING USING MATLAB OGE MARQUES Florida Atlantic University *IEEE IEEE PRESS WWILEY A JOHN WILEY & SONS, INC., PUBLICATION CONTENTS LIST OF FIGURES LIST OF TABLES FOREWORD
More informationAutomatic Aesthetic Photo-Rating System
Automatic Aesthetic Photo-Rating System Chen-Tai Kao chentai@stanford.edu Hsin-Fang Wu hfwu@stanford.edu Yen-Ting Liu eggegg@stanford.edu ABSTRACT Growing prevalence of smartphone makes photography easier
More informationA Fast Median Filter Using Decision Based Switching Filter & DCT Compression
A Fast Median Using Decision Based Switching & DCT Compression Er.Sakshi 1, Er.Navneet Bawa 2 1,2 Punjab Technical University, Amritsar College of Engineering & Technology, Department of Information Technology,
More informationPassive Image Forensic Method to detect Copy Move Forgery in Digital Images
IOSR Journal of Computer Engineering (IOSR-JCE) e-issn: 2278-0661, p- ISSN: 2278-8727Volume 16, Issue 2, Ver. XII (Mar-Apr. 2014), PP 96-104 Passive Image Forensic Method to detect Copy Move Forgery in
More informationAn Optimal Pixel-level Self-repairing Authentication. Method for Grayscale Images under a Minimax. Criterion of Distortion Reduction*
An Optimal Pixel-level Self-repairing Authentication Method for Grayscale Images under a Minimax Criterion of Distortion Reduction* Che-Wei Lee 1 and Wen-Hsiang Tsai 1, 2, 1 Department of Computer Science
More informationSecure Spread Spectrum Data Embedding and Extraction
Secure Spread Spectrum Data Embedding and Extraction Vaibhav Dhore 1, Pathan Md. Arfat 2 1 Professor, Department of Computer Engineering, RMD Sinhgad School of Engineering, University of Pune, India 2
More informationSplicing Forgery Exposure in Digital Image by Detecting Noise Discrepancies
International Journal of Computer and Communication Engineering, Vol. 4, No., January 25 Splicing Forgery Exposure in Digital Image by Detecting Noise Discrepancies Bo Liu and Chi-Man Pun Noise patterns
More informationA DEVELOPED UNSHARP MASKING METHOD FOR IMAGES CONTRAST ENHANCEMENT
2011 8th International Multi-Conference on Systems, Signals & Devices A DEVELOPED UNSHARP MASKING METHOD FOR IMAGES CONTRAST ENHANCEMENT Ahmed Zaafouri, Mounir Sayadi and Farhat Fnaiech SICISI Unit, ESSTT,
More informationCROSS-LAYER FEATURES IN CONVOLUTIONAL NEURAL NETWORKS FOR GENERIC CLASSIFICATION TASKS. Kuan-Chuan Peng and Tsuhan Chen
CROSS-LAYER FEATURES IN CONVOLUTIONAL NEURAL NETWORKS FOR GENERIC CLASSIFICATION TASKS Kuan-Chuan Peng and Tsuhan Chen Cornell University School of Electrical and Computer Engineering Ithaca, NY 14850
More informationImaging Sensor Noise as Digital X-Ray for Revealing Forgeries
Imaging Sensor Noise as Digital X-Ray for Revealing Forgeries Mo Chen, Jessica Fridrich, Jan Lukáš, and Miroslav Goljan Dept. of Electrical and Computer Engineering, SUNY Binghamton, Binghamton, NY 13902-6000,
More informationAutomatic Licenses Plate Recognition System
Automatic Licenses Plate Recognition System Garima R. Yadav Dept. of Electronics & Comm. Engineering Marathwada Institute of Technology, Aurangabad (Maharashtra), India yadavgarima08@gmail.com Prof. H.K.
More informationInformation Forensics: An Overview of the First Decade
Received March 8, 2013, accepted April 6, 2013, published May 10, 2013. Digital Object Identifier 10.1109/ACCESS.2013.2260814 Information Forensics: An Overview of the First Decade MATTHEW C. STAMM (MEMBER,
More informationIdentification of Bitmap Compression History: JPEG Detection and Quantizer Estimation
230 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 12, NO. 2, FEBRUARY 2003 Identification of Bitmap Compression History: JPEG Detection and Quantizer Estimation Zhigang Fan and Ricardo L. de Queiroz, Senior
More informationAn Automatic JPEG Ghost Detection Approach for Digital Image Forensics
An Automatic JPEG Ghost Detection Approach for Digital Image Forensics Sepideh Azarian-Pour Sharif University of Technology Tehran, 4588-89694, Iran Email: sepideazarian@gmailcom Massoud Babaie-Zadeh Sharif
More informationMotion Detector Using High Level Feature Extraction
Motion Detector Using High Level Feature Extraction Mohd Saifulnizam Zaharin 1, Norazlin Ibrahim 2 and Tengku Azahar Tuan Dir 3 Industrial Automation Department, Universiti Kuala Lumpur Malaysia France
More informationUM-Based Image Enhancement in Low-Light Situations
UM-Based Image Enhancement in Low-Light Situations SHWU-HUEY YEN * CHUN-HSIEN LIN HWEI-JEN LIN JUI-CHEN CHIEN Department of Computer Science and Information Engineering Tamkang University, 151 Ying-chuan
More informationHiding Image in Image by Five Modulus Method for Image Steganography
Hiding Image in Image by Five Modulus Method for Image Steganography Firas A. Jassim Abstract This paper is to create a practical steganographic implementation to hide color image (stego) inside another
More informationMLP for Adaptive Postprocessing Block-Coded Images
1450 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 10, NO. 8, DECEMBER 2000 MLP for Adaptive Postprocessing Block-Coded Images Guoping Qiu, Member, IEEE Abstract A new technique
More informationReversible Data Hiding in Encrypted color images by Reserving Room before Encryption with LSB Method
ISSN (e): 2250 3005 Vol, 04 Issue, 10 October 2014 International Journal of Computational Engineering Research (IJCER) Reversible Data Hiding in Encrypted color images by Reserving Room before Encryption
More informationNO-REFERENCE IMAGE BLUR ASSESSMENT USING MULTISCALE GRADIENT. Ming-Jun Chen and Alan C. Bovik
NO-REFERENCE IMAGE BLUR ASSESSMENT USING MULTISCALE GRADIENT Ming-Jun Chen and Alan C. Bovik Laboratory for Image and Video Engineering (LIVE), Department of Electrical & Computer Engineering, The University
More informationS SNR 10log. peak peak MSE. 1 MSE I i j
Noise Estimation Using Filtering and SVD for Image Tampering Detection U. M. Gokhale, Y.V.Joshi G.H.Raisoni Institute of Engineering and Technology for women, Nagpur Walchand College of Engineering, Sangli
More informationDeep Learning for Detecting Processing History of Images
Deep Learning for Detecting Processing History of Images Mehdi Boroumand and Jessica Fridrich, Department of ECE, SUNY Binghamton, NY, USA, {mboroum1,fridrich}@binghamton.edu Abstract Establishing the
More informationUNEQUAL POWER ALLOCATION FOR JPEG TRANSMISSION OVER MIMO SYSTEMS. Muhammad F. Sabir, Robert W. Heath Jr. and Alan C. Bovik
UNEQUAL POWER ALLOCATION FOR JPEG TRANSMISSION OVER MIMO SYSTEMS Muhammad F. Sabir, Robert W. Heath Jr. and Alan C. Bovik Department of Electrical and Computer Engineering, The University of Texas at Austin,
More informationLossless Image Watermarking for HDR Images Using Tone Mapping
IJCSNS International Journal of Computer Science and Network Security, VOL.13 No.5, May 2013 113 Lossless Image Watermarking for HDR Images Using Tone Mapping A.Nagurammal 1, T.Meyyappan 2 1 M. Phil Scholar
More informationA Spatial Mean and Median Filter For Noise Removal in Digital Images
A Spatial Mean and Median Filter For Noise Removal in Digital Images N.Rajesh Kumar 1, J.Uday Kumar 2 Associate Professor, Dept. of ECE, Jaya Prakash Narayan College of Engineering, Mahabubnagar, Telangana,
More informationTwo Improved Forensic Methods of Detecting Contrast Enhancement in Digital Images
Two Improved Forensic Methods of Detecting Contrast Enhancement in Digital Images Xufeng Lin, Xingjie Wei and Chang-Tsun Li Department of Computer Science, University of Warwick, Coventry, CV4 7AL, UK
More informationSelective Detail Enhanced Fusion with Photocropping
IJIRST International Journal for Innovative Research in Science & Technology Volume 1 Issue 11 April 2015 ISSN (online): 2349-6010 Selective Detail Enhanced Fusion with Photocropping Roopa Teena Johnson
More information